Further below, flagship essay:

Marriage as a moral disorder

but first

Two or Three Covenants in Place of Marriage


Across the western world there has been a torrent of controversy about same-sex marriage and associated rights to adopt children or, in some cases, to permit artificial procreation on behalf of same-sex couples. For all the passion and outrage on either side, little light has been thrown on the essential issue, which is that of the purpose, or rather purposes, of marriage. Once we call to mind some fairly obvious distinctions, things fall in place.

Marriage is a special kind of personal contract, but it is also a social institution. You might marry in near secrecy, but part of the essence of marriage is that the union is recognised publicly. This is the recognition that same-sex couples have craved and that traditionalists have wished to deny them. It is a separate matter to arrangements for procreating or bringing up children.

The heat of the debate has left one issue cold, which is the desirability of universal coupledom, or indeed whether this is seriously even conceivable. This failure is exemplified by the unfortunate slogan propagated in France of mariage pour tous ( = marriage for all) , which may sound more like a menace than a promise. As one kind of exclusion is ended, at least in terms of state recognition, so another kind appears, with all attention being focussed on the institution of coupledom (whether with or without a marriage certificate), as though this must be the universal norm. Those who live differently, whether by conviction or circumstance, are left outside. In point of fact, coupledom often comes at the expense of friendship, which is governed by loyalty rather than an exclusive — i.e. excluding — contract.


One purpose of marriage is to provide a robust institutional setting for bringing up children. It will not be disputed that children thrive better when there is constancy and continuity of attention from more than one adult. Two parents are better than one. It is argued below that two parents are not enough. Nor is it essential that these be biological parents.

A robust institutional setting for bringing up children must mean that those involved have rights and duties with regard to the child. The qualification “institutional” means that their involvement cannot be terminated on a whim. Outsiders — neutral, unknown and maybe anonymous ­ — must give their blessing to any change.

A separate set of purposes essential to marriage is the provision of emotional security and material solidarity between the persons of the couple. In short, a covenant to maintain a joint household.

"Marriage" cruelly conflates these two sets of purposes, and this failure to differentiate leads to no end of conflict, confusion and, of course, misery. It is for this reason that here the proposal is made to speak of covenants. Covenants resemble contracts, but they are less legalistic and, above all, give formal recognition to the role of the emotions and the virtue of loyalty.

One covenant that may be made is to bring up a child jointly. Much of the time, this will involve running a joint household. Normally, when the child has reached adulthood and independence, the covenant will have achieved its purpose and so have come to maturity — or fruition, its commitments having been fulfilled. Although long, it is temporary in nature.

A separate covenant that may be made is to share a household until “Death us do part” . This is not a flat-share or a business partnership. It is best exemplified by the rare but poignant cases where two siblings spend a lifetime together in a single household, presumably without any sexual relationship between them, although this is of no concern to anyone outside.

The household here to be shared in permanence also stands for solidarity, as “in sickness and in health”, or indeed “for richer or poorer,” while legal recognition involves rights & duties in cases where one of the couple is incapacitated.

A problem may arise when one of the couple dies, but only because, in most jurisdictions, of the onerous nature of inheritance tax. (The revenue authorities may de facto confiscate the jointly held home, leaving the remaining person destitute.) That is, the problem would seem to arise solely from a heavy-handed, non-discretionary tax bureaucracy interfering with common-law property rights. (The non-discretionary aspect is associated with the contemporary prioritisation of equality; strict equality being an unobtainable goal and so always at odds with discretion, but that is another subject.)

A household covenant might more usually be equivalent to a marriage as commonly understood when no children can come naturally and no adoptions are planned.

Obviously two people might resolve to enter into both covenants, though this might be rash since one cannot in early adulthood (i.e. when most fertile) know the person one will be, much less so the person the other will become, a score of years later. This would be more like entering a convent than a covenant.

Indeed, in public reflection about the institution of marriage, much light might come from considering in parallel the institution of celibacy such as is exemplified when someone takes a vow of celibacy, which is not only a promise not to marry, but also a vow of chastity. Would it be possible to go to a government office to register this vow? And if, subsequently, one broke it, would there be legal recourse for the breach of vow to be punished? Or would one be able, legally, to register a revocation?

These reflections demonstrate that marriage is first & foremost a religious sacrament. One does not have to be religious to acknowledge this. A marriage — proper — takes place in the eyes of God. So an atheist cannot marry. An atheist may, however, very well enter into a solemn promise — a binding agreement — to care jointly with another adult for a child, or else to share a household until “death they do part”.

Hence it is impossible for the state to conduct marriage ceremonies, between whomsoever. And consequently impossible for the state to regulate marriage. The tasks of the state here are, rather, to register (i.e. acknowledge & anchor) covenants for the upbringing of a child or the maintenance of a life-long household. Much would be gained if the state were to distinguish these two tasks and make proper and separate provision for them.


There is a third — rather obvious — purpose of traditional marriage, and that is to provide the spouses mutually with a permanent sexual partner. The Enlightenment philosopher Immanuel Kant maintained — notoriously — that marriage is a contract for the exclusive use of the sexual organs of the other. A more realistic and liberal version for the contemporary world would be that it involves a promise of availability, or a granting of priority.

In France there was a case recently of a wife being awarded compensation because her husband had failed to fulfil his conjugal duties. Although in the reported case the husband’s neglect seems to have been a matter of his indifference, in other cases the withholding of sex is used as an instrument of power in the dynamics of a couple. On the other hand, progressive opinion has led to legal recognition of the possibility of rape within marriage, a possibility that some traditionalists contest.

It will be obvious that no brief treatment of these conflicting perspectives can begin to do justice to the enormous range of situations that can occur. The question arises, therefore, whether, in this regard, marriage can be given any kind of state recognition. Again, it would seem that marriage is a religious sacrament, or covenant, which the adherents to the faith must define and police on their own, independently of any legal involvement.

There is, however, another institution that has been hitherto religious, or rather Christian, and whose recognition and regulation by the state could be universally beneficial.


Godparenthood: Re‑weaving the social fabric

Granted it is better that a child should have more than one adult to bond to, some of the invective against parenthood in a same-sex marriage would seem vindictive. It would amount to depriving some children of a good. But, if we presume, to be conventionally on the safe side, that it is best for a growing child to have adults of both sexes to bond to, there is a simple solution, which may be seen either as a compromise or an ideal. It may also present a safety valve against gross mistreatment.

There is an ancient institution to provide additional assurance that children are cared for, roughly, as they should be, and that is godparenthood. All we need do is give this institution legal recognition, such that parents and especially any single parent be urged to nominate two or three godparents, who would have legally enforceable rights and obligations. That is, a godmother and a godfather (not necessarily in any liaison) would be expected and empowered to visit their godchild regularly, for example once a fortnight, and held partly responsible (with penalties) if the child came to serious harm.

They might be grandparents, or aunts & uncles, but ideally they would, as originally in medieval Christendom, be unrelated persons who lived nearby. This mechanism could work wonders in establishing or reinforcing social cohesion. It would also encourage the institution of long-term friendship, which suffers grievously when everywhere all emotional eggs are invested in the excluding relationship of marriage.

Such a rejuvenated and legally enforceable institution of godparenthood would calm the passions in the contentious debate about gay adoption. If two men, or two women, are bringing up a child jointly, then let them appoint godparents of the opposite sex to provide some gender balance in their child’s growing perception of the human world, it being understood that we understand gender imperfectly, and should err on the side of balance.

Although originally an ecclesiastical institution, there is no need for godparenthood to have religious undertones.


Since some marriages undoubtedly succeed, why is it that others fail? Is this just a matter of individual and all-too-human shortcomings? Or is it possible that marriage, when held up as a universal moral order, has failings of its own? Would it actually be possible, if only we were all mature enough, for everyone to be a partner to a happy, exclusive and lifelong relationship of sexual and emotional intimacy? Or could it be that, both within and without marriage, the players who lose out are merely the victims of the rules of the game, that some, at any rate, are really just well-rounded persons who fail to fit into the square hole of marriage?

Changed circumstances

Before examining the concept of marriage as such and the values it assumes, let us look at some assumptions and clear the ground of related issues which are only sometimes relevant.

Our times have seen sweeping changes in the material circumstances governing our sexual and affective lives. These are:

* Control over fertility. Connected with this is the desirability (in the light of high population levels and high per capita consumption of natural resources) of a low birth rate, and the fact that people no longer need provide for their old age by having children themselves.

* Acceptance of the principle and partial realisation of substantial equality between the sexes. Connected with this is the material feasibility of single-person households, with our affluence offering us technical convenience unimagined in previous epochs.

These changes (i.e. the attenuation of certain constraints) mean that age-old structures and assumptions no longer hold. While a traditionalist might see in the system of marriage the wisdom of the ages, we have meantime entered veritably a new era.

Definition of marriage

By "Marriage" I mean monogamous marriage and marriage-like arrangements, e.g. common-law marriages; i.e. much of the analysis also applies to "relationships". The principal elements involved in marriage — its defining characteristics — are:

* A sexually exclusive relationship — this is the crux of the matter; "open marriages" are outlawed.

* A single focal relationship, overriding other claims of friendship. The marriage partner is the most important person is the life of the other, both day-to-day and long-term.

* Co-habitation. Married people share a household

* The intention of permanence

* Solidarity — notably in times of hardship, sickness and in old age.

The important non-essential element is joint upbringing of children issuing from the marriage. First some comments on this last aspect.

Marriage and parenthood

In appropriate material circumstances, there is no compelling reason why the responsibilities of parenthood must be exercised within marriage, i.e. in a joint household and with a continuing, sexually and emotionally intimate relationship between the parents. All that is required, in practical terms, is that the two households be located conveniently close to each other and that the parents should have a friendly working relationship. (The key material circumstances involved here are flexible working hours or else material independence and, crucially, the availability of suitable housing. It is incidentally apparent that, as a society, we have come to place more emphasis on providing a variety of other goods and services, including many that are doubtless trivial, rather than designing our living and working environments to be adaptive to the lives of the affections.)

It is important to stress that other considerations apply when material circumstances fall short of what is required. Children need stable relationships with adults of both sexes, and the primary obligation to provide those relationships rests with the child's mother and father, often overriding their other duties. (This does not mean that the two child-parent relationships are enough; children need a multiplicity of relationships with adults of both sexes, both in order to correct for the idiosyncrasies of their parents and to provide substitutes in the event of tragic separation.)

Even if the above principle is not accepted (i.e. that parenthood can be adequately exercised by parents in separate households), there is the further point that, in an average life span of over seventy years, of which fifty are spent in full adulthood, care for children should not normally be taking up more than, say, twenty years. There is no reason why the life style for the other half of adult life should be dictated by considerations appropriate, if at all, only to this special twenty-year period.

Practical arguments for sexual fidelity

Traditionally the social and economic need for fathers to take their share of responsibility for any children produced by a sexual relationship has justified a rule of sexual fidelity. But note that this justification was even then only watertight as long as the nature of the sexual intimacy was liable to actually produce children. Today we have considerable control of fertility, with the result that this argument, while retaining some force, is much diminished.

Personal expectations and demands

There have been major changes in what is expected of a relationship. Since marriage is no longer an economic necessity, the focus has shifted from material considerations to personal qualities. A good husband, a good wife, are no longer defined in terms of the fulfilment of specific roles such as being a reliable provider or an efficient housekeeper. What most of us now expect from a prospective partner is defined largely in terms of emotions and personalities, and is hence highly individualistic. Often it is even in terms of shared leisure interests. Typically, enormously complex demands are now made of a prospective marriage partner. It would be too easy simply to censure people for making these demands. If one takes the notion of marriage as a life-long (and sexually exclusive!) union at all seriously, then it would seem to be rather important to set higher criteria than those for a simple friendship.

The rationale of fidelity

If monogamy is claimed as the only justifiable setting for sexual intimacy, there must be a rationale for this. At a general level, it can be agreed that humans and human societies have a continuing need (i.e. independent of considerations related to fertility) to integrate sexual behaviour into the life of the emotions. That is, we try to make sexual intimacy expressive of the mutual emotions felt by those involved. More precisely, we say that those emotions should be ones of affection, fondness, liking or love. So far, so good. ###> AIDS: In the meantime we have, again, as in some previous centuries, a fatal, sexually transmitted disease. This is a powerful argument for hygiene. It is only a weak argument for fidelity. <#### In this connection, there is a general question of how risks associated with infectious diseases should be handled. Always avoided at all costs? This would seem to be too extreme. There are sometimes good reasons for taking small risks.1 Furthermore, we might bear in mind the way we, as individuals and as a (world) society, handle other risks. It needs also to be said that, although the risk may now be nearly everywhere, it is not remotely everywhere equal.

The problem arises because of the exclusive nature of the claims, and most notably the sexual claims, inherent in marriage. Presumably the justification for this exclusivity is that the marriage relationship is (ideally) the central relationship in a person's life, with a much more intense helping of affection and love than is to be found in other (sexually attractive and socially tolerated) relationships. Sexual intimacy is then seen as expressive of the uniqueness of the relationship; extra-marital sexual relationships are ruled as undesirable because they belie or undermine the centrality of the marriage relationship (legitimate sexual intimacy being interpreted as always expressive of central emotions).

One assumption here is that a person needs (and therefore should ideally live within) a single central relationship. But it is difficult to see what an argument for this claim would look like. It might be an empirical claim about what people naturally prefer when they are free to choose, but this is a highly theoretical circumstance. Apart from the commonplace that people never grow up within a vacuum, there is the consideration that individuals do not usually have anything like a practical choice in the matter, any more than they have a choice about whether or not to enter a contract of employment of some kind.

Perhaps the argument for monogamy draws its strength from a claim about what kind of arrangement generates the most potential for happiness, fulfilment and the like. Here it would be easy to provide apparent counter- examples. But the claim is too vast for anecdotal matter to decide the issue. Each of us may form a conviction one way or the other; in the final analysis, it seems more like a matter of faith than one for demonstration.

Assuming that sexual intimacy is, ideally, expressive of love, it needs to be demonstrated that it is only possible to love a single person at any one time.3 This is counter-intuitive. It might, however, be possible to redefine love in a strong sense such that intense love is always focused on only one person. We might genuinely wonder, for instance, whether it is possible to fall in love, or be in love, with more than one person at a time. But we might equally question whether such a peak of intensity of feeling can be maintained for long, let alone over a lifetime. If it is not maintained, then there is a justification for either later terminating the sexual intimacy involved in the relationship, or else extending sexual intimacy to a new love.

For the present defence of fidelity to hold, the love ideally present or sought in marriage has, we have seen, to be either exclusive by definition or else must be a love of such intensity that it practically rules out comparable depth of feeling towards anyone else. Now there is no reason to suppose that life will necessarily present us with a person whom we feel bound to quite this closely and who also, felicitously, returns the feeling in equal measure. It will happen in some lives, fail to materialise in others, and in yet others there will be an essentially one-way affection, at least of this intensity. The move now open to the defence of fidelity is to claim that the special love involved is something which has to be worked at, that it is not a love which just happens.

Arguably no love just happens; the new affection is tended with hopes and meetings until it is sometimes suddenly — happily or tragically — out of control. Neither can love, as commonly understood, be forced. Yet for their argument to hold, the advocate of fidelity has to redefine love as something that can substantially be controlled, i.e. subjected to the will of the individual. This redefinition is implausible. The common word “love” involves necessarily a strong affective element and hence a certain spontaneity; love is not invested like trust in a business partnership.

The advocate of fidelity must now move the focus to the marriage relationship as such. The relationship, rather than the sense of love, must be nurtured until, with the years, it has become something special or irreplaceable, i.e. until each partner has become the most important person in the life of the other. "For better or worse," I should sceptically add, for familiarity is not the same as love.

This is the nub of the problem. The advocate of fidelity is eventually forced to discard the centrality of love, at least in any ordinary sense, and replace it with some such notion as long-term commitment. But this makes the connection of the relationship specifically with sexual intimacy ever more tenuous and artificial. The truth of the contention becomes clearer and clearer: the relationship becomes special because it has been made the unique arena for sexual intimacy, while the justification for restricting sexual intimacy to this one relationship is that it is special. That is, a wholly vacuous specialness is manufactured. In other words, there is in principle also no moral reason (there may still be prudential ones) for restricting sexual intimacy to a single relationship.

Mismatches: the non-universalisability of the marriage ideal

In the long term, sexual desires can rarely be ignored with impunity. The following argument therefore assumes that the great majority of people need sexual connection of some kind and that it is desirable that they should find some sexual fulfilment. The corollary is that celibacy imposed by force of circumstance is an evil.

Just now the point was made that not everyone is likely to meet the person whom they can relate to and love unreservedly and who also, felicitously, returns the sentiment. It was argued that this fact would force the advocate of fidelity to redefine love so radically as to lead eventually to the collapse of his argument.

Now in forming and developing relationships a great deal always turns on just how much compatibility and closeness is demanded. This said, the advocate of sexual exclusivity needs, for their argument to carry any conviction, rather a high level of compatibility and closeness. However, once expectations and demands begin to rise only a little, people encounter difficulty in finding the suitable partner or else they seem to find themselves with the wrong partner. Where does the fault lie, in the people who ask too much, or in the value system that calls for such a high investment in a single relationship?

If the justification for sexual exclusivity is sought in the specialness of the relationship (or else in the specialness of the other person), then we must ask whether everyone can realistically have such a relationship if they so choose. Is this statistically feasible? It would certainly be conceivable for nearly everyone to be a partner in an exclusive relationship of sorts. And equally conceivable for this relationship to be a sexually intimate one excluding third parties. But such partnerships cannot reasonably be expected also to be ones of a meeting of minds, of wide-ranging compatibility, of depth of companionship. Yet once these features are absent, again, the moral rationale for fidelity collapses.

The point is crucial, and I will restate it — twice. Assuming affections may be coaxed but not manufactured, and since affections are not automatically mutual or automatically associated with sexual attraction, it is statistically probable that many people will be unmatched or ill-matched. If they are ill-matched, a mockery is made of the whole rationale of sexual intimacy being expressive of the special closeness of the marriage partners. There is no longer anything inherently unique or especially valuable about the relationship, only the uniqueness imposed externally by the norm of exclusivity in sexual matters. The justification for keeping the sexual relationship exclusive has to be that the relationship is on other grounds and in other respects unique and special such that sexual intimacy and exclusion can mirror this unique and special character. But now, in practice, the only thing special and unique about many of the relationships is that they involve sexual intimacy.

This problem only arises as long as the rule of sexual fidelity, as a moral (rather than hygienic or progenitory) rule, is propagated as binding on all. The ideal of exclusive love (in an ordinary and substantial sense of the word love) as the only proper setting for sexual intimacy cannot realistically be universalised. This does not mean that it cannot be realised some of the time. (It could theoretically be universal in a world which were governed by divine providence and guardian angels, but it cannot be universal in a world subject to the — creative? — chaos of non-mutual and conflicting attractions.)

Here the second restatement of this crucial point, this time in statistical terms. Imagine a model of people forming couples. Let us suppose that the average woman in the realm is three inches shorter than the average man. Let the tyrant decree that each woman is to marry a man who is three inches taller. We intuitively see that havoc would result, at least in the absence of centralised matching by computer. Our intuitive insight could doubtless be demonstrated by mathematical modelling using probability theory.

Let us now add just one more variable. Suppose the average woman is three years older than the average man and that the tyrant issues a further decree that each woman must marry a man three years her junior. With just two variables, the headache is complete. Yet in the area of personal relationships we are dealing with a veritable multitude of variables.

The ideal and the good

Let us suppose for a moment that a close, lifetime relationship, which is the ideal of marriage, really is a most desirable destiny for all. (This is a big supposition. The institution of marriage lends legitimacy to the staking of exclusive claims, and hence to possessiveness and jealousy. The latter can scarcely count as virtues.)

Let us suppose furthermore that, notwithstanding the foregoing argument, it could reasonably be achieved by everyone, i.e. that there were no systemic problems about its universalisability. This ethic still runs foul of another difficulty, namely that setting something up as an ideal to be pursued by all is likely to encourage falseness in one shape or another. For example, assuming that sexuality constitutes a powerful drive, if it is only acceptable to exercise that drive in the presence of certain rather high-flown emotions and attitudes, people will consciously or unconsciously fabricate the appropriate feelings. Those feelings, even if they are not wholly counterfeit, will consequently lack staying power.

In other words, once an ideal is firmly in place, people will feel constrained to comply, and be it at the cost of suppressing (either publicly or psychologically) what they really feel and want. In the present case, if it is put about that people (and notably adolescents) can only have sex when what they feel for each other is love (rather than, say, simply affection), then they will either pretend to such feelings or else — and I suspect this is far worse and widespread than straightforward deceit — they will manufacture the said feelings. In consequence, they will likely be misled into confusing sexual desire with love.

Now it might seem that this blending of the two (desire and love) is the ideal state, indeed just what we have been striving for, and in a sense it is. But do we achieve the real thing by suppressing the distinction? Or do we not rather introduce confusion into the hearts of the best, while fostering hypocrisy in the rest? The word love is easily spoken, and it is only a little more difficult to imagine that it has been achieved. Especially if you have the incentive of socially approved sex to reward you.

The general fallacy at work here might be described as the confusion of the ideal with the good.

There is a crucial difference between, on the one hand, a statement that sexual intimacy can be expressive of love (and when it is so expressive this fortuitous event is to be welcomed), and, on the other hand, the setting up of an ideal unity between desire and love, which all should strive for. There is a presumption here even that we have the power to choose to achieve this harmony. This presumption is, I contend, a form of hubris. Or, in a more modern, a Wittgensteinian idiom, it is what happens when thought is idling. Too often, indeed, ideals are but the product of idle minds. What we should instead require, of ourselves and of others, is behaviour that is good enough.


First the argument from common experience tells us how difficult it is to find and win someone with whom we can forge more than a limited partnership. Should we then lower our sights? But if it does not much matter whom you marry as long as a rough and ready compatibility is ensured, then equally it ceases to matter that you should be faithful. Unless the argument for sexual exclusiveness is to be one of fidelity for its own sake.

The empirical premise of this argument, namely the practical difficulty people face when seeking a suitable partner, was then shown to be a nearly necessary truth. That is, the statistical probability of no or few placement difficulties arising is negligible. Hence, the failure of marriage as a social system is pre-programmed, and does not generally reflect personal failings. (This argument from non-universalisability does not presuppose acceptance of a Kantian standpoint, although a Kantian would have to accept it.)

Finally, a separate argument distinguished between a set of circumstances being good in the sense of felicitous and these circumstances being an appropriate target for acts of will. We should not seek to control and determine everything, because the attempt is likely to fail and, indeed, create more havoc than letting things be. In matters connected with sexuality, it is arguably wholly inappropriate, for spontaneity is of the essence here, the nature of the erotic experience being a letting go, an abandon, and so a relinquishment of control.


The following PDFs and introductory text are previous versions dating from long ago. The above texts are revisions in early 2022 with some material deleted, times having moved on.

Two or Three Covenants in Place of Marriage
followed by
Godparenthood: Re-weaving the social fabric
1650 words

> Download large-font PDF for easy reading on mobile devices.


Marriage as a moral disorder
Why monogamy, as a universal system, must fail

Since some marriages seemingly succeed, why is it that others fail? Is this just a matter of individual and all-too-human shortcomings? Or is it possible that marriage, when held up as a universal moral order, has failings of its own? Would it actually be possible, if only we were all mature enough (what an odd idea!), for everyone to be a partner to a happy, exclusive and lifelong relationship of sexual and emotional intimacy? Or could it be that, both within and without marriage, the players who lose out are merely the victims of the rules of the game, that some, at any rate, are really just well-rounded persons who fail to fit into the square hole of marriage?

> Download the whole essay as PDF.